LINGUIST List 28.430

Fri Jan 20 2017

Review: Linguistic Theories; Morphology; Syntax: Sims (2015)

Editor for this issue: Clare Harshey <clarelinguistlist.org>


Date: 13-Jul-2016
From: Daniel Walter <dwalterandrew.cmu.edu>
Subject: Inflectional Defectiveness
E-mail this message to a friend

Discuss this message


Book announced at http://linguistlist.org/issues/27/27-957.html

AUTHOR: Andrea D. Sims
TITLE: Inflectional Defectiveness
SERIES TITLE: Cambridge Studies in Linguistics
PUBLISHER: Cambridge University Press
YEAR: 2015

REVIEWER: Daniel Walter, Emory University

Reviews Editor: Helen Aristar-Dry

SUMMARY

“Inflectional Defectiveness” by Andrea D. Sims explores the linguistic phenomenon of paradigmatic gaps and investigates what perpetuates their existence in and across languages. Sims rejects the notion that defective patterns in morphemic paradigms are marginal, epiphenomenal failures of the underlying generative language system. In place of such a theory, she focuses on how inflectional defectiveness is actually a result of the same historical, social, and cognitive factors that drive other morphological patterns. The book is divided into eight chapters and includes analyses of multiple languages to provide evidence for the author’s claims, with the bulk of the evidence and analysis focusing on Modern Greek and Russian.

The first chapter is presented as an introduction to inflectional defectiveness. The author begins by outlining the seemingly automatic, productive nature of morphological paradigms in English. She opens with the example of the word “Google” and how people have applied English morphemes to allow for its use in different contexts (e.g. googles, googled, googling, etc.). If language users apply these patterns so readily, the question then arises, why do paradigmatic gaps exist? The chapter then lays out three ways of viewing these gaps: “random anomalies, epiphenomena, or (almost) normal morphological objects,” (Sims, 2015, pp. 7). First, the random anomaly stance posits that these gaps exist outside of the core functions of grammar and thus are not part of normal grammatical processes. Second, the epiphenomenon position views paradigmatic gaps as the result of conflicting morphological feature specifications. Third and finally, the normal morphological object view holds paradigmatic gaps as accountable to the same processes that govern other morphological objects, which is the viewpoint taken in this book. The author concludes by outlining her position on the structure of the lexicon, which she does through a comparison of formal, inferential-realizational and psycholinguistic models. Her starting point for her inquiry is a “dynamic, paradigmatic structure of the lexicon”, which operates from a position “at the nexus between formal morphology, linguistic typology, and psycholinguistics,” (Sims, 2015).

The second chapter is devoted to defining the scope of inflectional defectiveness and teasing it apart from other linguistic phenomena, including periphrasis. The author provides a working definition in logic-based terms:

“a. IF there exists a set of morphosyntactic and/or morphosemantic feature values F that is well-defined and morphologically encoded for at least one lexeme belonging to part of speech C;
b. AND IF there exists a well-formed syntactic structures S that requires F in combination with some lexeme L belonging to C;
c. BUT any form of L(subC) that is inserted into S procures an ungrammatical construction;
d. THEN the paradigm cell defined by {L(SubC),[F]} is defective.”

In prose, this equates to the idea that if a word, as part of a particular part of speech, has some strongly expected morphological paradigm, which it should follow, but does not, then that intersection of morphological paradigm and word is seen as defective. In order to justify this distinction, the author posits different syntactic and morphological levels, in which the morphological level does not have to directly map onto the syntactic. In other words, although the syntactic level may require a complete paradigm, the morphological level does not and thus a mismatch can occur. This leads to what the author describes as a “fuzzy boundary” and a resulting “gradient” view of defectiveness. Ultimately, cross-linguistic differences between languages in how well-defined a certain set of morphological features is, the distributional patterns of those features, and other linguistic external and internal factors affect the stability of the morphological paradigm as posited in part “a” above. In sum, the definition advanced in this chapter is reliant on the particulars of the morphological paradigms evidenced in a specific language and its usage.

Chapter Three is primarily focused on the causes of paradigmatic gaps, to wit the author offers four: a lack of semantic or pragmatic need, morphology-phonology interface issues, the application of multiple morphological forms, and problems at the level of the morphosyntactic structure. Gaps related to a lack of semantic and pragmatic need are said to arise from missing contexts in which forms would actually be utilized. The morphology-phonology interface issue deals with gaps that result from morphological patterns being mapped onto difficult surface phonemes. The issue regarding morphological forms is related to the problem when multiple morphological rules compete for application. Lastly, morphosyntactic issues are argued to result from changes in the underlying features of a language, such as the loss of particular cases. In addition to these four causes, the author provides evidence from the Witsuwit’en language that shows there is sometimes no discernible cause for these patterns of defectiveness. The author concludes her argument by positing that these causes are necessary but not sufficient to explain patterns of defectiveness.
Chapter Four focuses on the competing outcomes of syncretism and defectiveness when the morphological system does not completely overlap with syntactic specifications. The author points out that this competition at the morphological level of language can result in three possible outcomes: syncretism overrides defectiveness, defectiveness overrides syncretism, and defectiveness follows the pattern of syncretism. The author’s main point is that the interrelation between defectiveness and syncretism show lexically specific defectiveness that appears as part of the inflection system because, as she argues, the syntactic and morphological paradigm levels can have “distinct shapes”. Of necessary mention is the author’s note that real-world, empirical data pose “a significant challenge to a theoretical account of defectiveness,” which leads her turn to other functionalist, connectionist, and emergentist approaches.

Chapter Five turns from descriptive characteristics of paradigmatic gaps and takes a closer look at how they can arise from the same principles that organize the entire inflectional system. For this, the author uses examples from nominal inflectional patterns in Greek, which exhibits defectiveness in the genitive plural forms, as well as some genitive singular forms. The results from the Greek data indicate that a lack of paradigmatic cohesion is one of the primary reasons for the gaps in these paradigms. The author concludes the chapter by drawing into question whether paradigms themselves should be the focus of our attention or paradigmatic relations.

Chapter Six builds on the idea presented in Chapter Five, that defectiveness and paradigmatic gaps are themselves morphological objects. One idea the author attempts to counter is that lack of form does not provide evidence for language learners. The author argues that while a lack of morphology would be a problem if viewed in isolation, the fact is that these gaps develop within systems that have morphological forms and that these gaps can be viewed as signals of morphology because of the evidence and expectations provided by other paradigms in the language. Using statistical analyses of Greek corpora, the author points to multiple causes for these gaps, which include a lack of cohesion in the paradigm and speakers’ ability to avoid certain forms through the use of prepositional phrases. In addition, she includes data from elicitation and ratings studies showing that speakers have reanalyzed some of these patterns as lexeme-specific.

Chapter Seven brings the previous chapters to a head by investigating the problem of learnability and the dynamic organization of the lexicon. Using data from Russian first person singular gaps, the author shows that no single explanation seems to be able to capture the defectiveness exhibited in the language. The author argues that for these gaps to exist they must be learnable by the same principles as other aspects of grammar. Using Bayesian modelling, the author shows how absences can be inferenced and seen as morphological objects by learners. Morphological, phonological, and semantic patterns that emerge dynamically from usage form lexical gangs that reinforce particular patterns. The author argues that these lexical gangs and learning processes overcome the problems of negative evidence, sparse data, and minority patterns.

Chapter Eight summarizes the previous seven chapters and focuses on reinforcing the author’s claim that defectiveness and the resulting gaps are simply allomorphs. The chapter concludes with a critique of linguistic theories which treat inflectional paradigms as either regular or irregular, based on the information that irregular patterns, and regular for that matter, vary in conformity from one word and one language to the next, and that these patterns result from common learning and historical processes.

EVALUATION

In “Inflection Defectivenss”, Andrea Sims shines a light on a specific, controversial, problematic, and therefore often ignored aspect of many of the world’s languages. In doing so, the author describes a number of paradigmatic gaps in multiple languages.

The initial impression given by the first few chapters is that the book is written from a formalist perspective. A number of the items are laid out in a manner reminiscent of Chomskian, logic-based approaches to the study of language. For readers not familiar with somewhat esoteric linguistic terminology, accessing the first few chapters may be a challenge. In addition, some of the explanations of particular linguistic problems focus on forms, rather than cognitive explanations for the appearance of those forms.

What is interesting is that, as the book goes on, the author slowly changes the feel and focus of her argument. As she tries to bridge the gap between formalist linguistic and cognitive approaches, more and more material, methods, and perspectives from methods unfamiliar to formal linguists appear. For readers expecting a purely formal linguistic text, there will be certain areas in which additional reading in connectionism and cognitive linguistics may be necessary. Some particularly helpful articles to read in conjunction with this text would be Rumelhart and McClelland (1987) for Parallel Distributed Processing (PDP), MacWhinney, Leinbach, and Taraban (1989) for Competition Model and Cue-based learning, and Chater, Tenenbaum, and Yuille (2006) for Probabilistic and Bayesian Modeling approaches.

Because the author tries to reconcile two often opposing views on the nature of language, readers from both camps will likely find points of contention with this book. On the one hand, cognitivists may find some of the explanations, particularly in the beginning chapters, overly rule-based. On the other hand, formalists may take issue with the way the author begins to conflate language form and use to explain problems of learnability, and moves away from formal linguistic explanations. In sum, the author takes on the heavy task of bringing two theoretically opposed camps together in order to explain the data, and she accomplishes it to a certain extent.
This book is intended for advanced linguistic researchers (graduate students and above) who are interested in morphology. A significant amount of prior knowledge across the linguistic spectrum is necessary for comprehension, and even then additional readings in particular areas may be necessary.

REFERENCES

Chater, N., Tenenbaum, J. B., & Yuille, A. (2006). Probabilistic models of cognition: Conceptual foundations. Trends in cognitive sciences, 10(7), 287-291.

MacWhinney, B., Leinbach, J., Taraban, R., & McDonald, J. (1989). Language learning: Cues or rules? Journal of Memory and language, 28(3), 255-277.

Rumelhart, D. E., & McClelland, J. L. (1987). Learning the past tenses of English verbs: Implicit rules or parallel distributed processing. Mechanisms of language acquisition, 195-248.


ABOUT THE REVIEWER

Dan Walter is a Visiting Assistant Professor of German and Writing at Emory University Oxford Campus. His research interests include Second Language Acquisition, Morphosyntax and Sociocognitive approaches to language learning. He received his PhD from Carnegie Mellon University in Second Language Acquisition in 2015.

Page Updated: 20-Jan-2017